GCI Results

Statistics For Tropical Mesoscale Convective Systems Derived From Global Cloud Imagery

Mesoscale convective systems in the tropics have been identified, tracked and statistical distributions computed using Global Cloud Imagery (GCI) (Salby et al, 1991; Tanaka et al, 1991). Two different feature detection schemes have been used:

(a) a scheme based on the shape of cloud clusters identified in the thresholded images, and using distance transforms to identify suitable points to track.

(b) a scheme based on smoothing the cloud top temperature over each cloud cluster and determining minima in the cloud top temperature.

The GCI data is currently only available for July83-June84, so both schemes have been used on the JAS83 and DJF8384 periods. The statsitics show that both schemes produce resonably similar results, although I hope to do some more analysis on this.

Statistics for JAS-SHAPE

GCI_JAS83_SHAPE_PART1.ps.gz

GCI_JAS83_SHAPE_PART2.ps.gz

Statistics for JAS-SMOOTH

GCI_JAS83_SMOOTH_PART1.ps.gz

GCI_JAS83_SMOOTH_PART2.ps.gz

Statistics for DJF-SHAPE

GCI_DJF8384_SHAPE_PART1.ps.gz

GCI_DJF8384_SHAPE_PART2.ps.gz

Statistics for DJF-SMOOTH

GCI_DJF8384_SMOOTH_PART1.ps.gz

GCI_DJF8384_SMOOTH_PART2.ps.gz

Thanks to Murry Salby and Jackie Gratrix for supplying the data for this work.

Salby, M. L., H. H. Hendon, K. Woodberry, and K. Tanaka, 1991: Analysis of Global Cloud Imagery From Multiple Satellites, Bull. Am. Met. Soc., v72, pp 467-480.

Tanaka, K., K. Woodberry, H. Hendon, and M. Salby, 1991: Assimilation Of Global Cloud Imagery From Multiple Satellites, J. Atmos. Oceanic Tech., v8, pp 613-626.